Skip to main content
Conversational AI Engine: Create advanced voice apps by combining Agora's real-time audio with powerful conversational AI.

Deepgram

Deepgram provides fast, accurate automatic speech recognition with advanced AI models optimized for real-time streaming and conversational applications across multiple languages.

Use a preset​

To use Deepgram ASR with an Agora managed key, specify one of the following presets in the preset field when starting an agent:

  • deepgram_nova_2
  • deepgram_nova_3

When using a preset, you do not need to provide the API key, endpoint URL, or model. You can still use the asr field to configure additional settings such as language and keyterm. To provide your own API key, see Sample configuration.

Sample configuration​

The following example shows a starting asr parameter configuration you can use when you Start a conversational AI agent.


_9
"name": "unique_name",
_9
"preset": "deepgram_nova_3",
_9
"properties": {
_9
// ...
_9
"asr": {
_9
"language": "en-US",
_9
"keyterm": "term1%20term2"
_9
}
_9
}

Key parameters​

paramsrequired
  • url stringnullable

    The WebSocket URL for Deepgram's streaming API.

  • key stringrequired

    The API key used for authentication. Get your API key from the Deepgram Console.

  • model stringnullable

    The speech recognition model to use.

  • language stringnullable

    The language code for speech recognition (For example, en, es, fr). See supported languages for language codes.

  • keyterm stringnullable

    Boost specialized terms and brands. Only compatible with the nova-3 model.

The parameters listed on this page are validated for use with Conversational AI Engine. Required parameters must be provided as documented. Any additional parameters are passed through directly to the underlying vendor without validation. For advanced configuration options, model selection, and detailed parameter descriptions, see the Deepgram API documentation.

caution

The following parameters are not passed through and instead use default values set by Agora:

  • callback
  • callback_method
  • channels
  • encoding
  • multichannel
  • sample_rate